Newest papers

Gate3DGen: A Pre-processing Model that Selectively Purifies Real-World Images for 3D Generation

Xiaolin Li, Zihao Huang, Fanpeng Meng, Chen Yang, Guanjun Wu, Kai Zhang, Jiemin Fang, Qi Tian · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31649 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

La generación de mesas 3D de imagen única ha logrado una alta fidelidad en los parámetros de referencia curados, pero degrada drásticamente en las imágenes del mundo real con iluminación desigual, borrosa y ocluciones. Purificar tales entradas requiere decidir cuándo interferir y preservar las pistas visuales necesarias para la generación 3D, mientras que deja las entradas limpias sin cambiar. Introducimos Gate3DGen, un modelo de preprocesamiento selectivo orientado a un paso que lleva las entradas de imagen a los generadores 3D existentes sin retraerlas. Un router consciente de la degradación oculta las entradas clasificadas como limpias y extrae características para guiar la purificación de las entradas restantes. Condicionado a estas características, un purificador de un solo paso

Machine translation (Meta M2M100); the English original follows.

Abstract

Single-image 3D generation has achieved high fidelity on curated benchmarks, yet degrades sharply on real-world images with uneven illumination, blur, and occlusions. Purifying such inputs requires deciding when to intervene and preserving the visual cues needed for 3D generation, while leaving clean inputs unchanged. We introduce Gate3DGen, a selective pre-processing model that gates image inputs to existing 3D generators without retraining them. A degradation-aware router bypasses inputs classified as clean and extracts features to guide purification of the remaining inputs. Conditioned on these features, a 3D-oriented one-step purifier purifies degraded inputs with spatial losses that encourage preservation of object identity, contours, and textural cues. To provide paired supervision, we construct a synthetic data engine that renders clean views of 3D assets and synthesises composite degradations. On the synthetic benchmark, the purifier improves image fidelity and downstream 3D geometry in only 2.3s on average, reducing TRELLIS.2 mesh distance by 61.5% relative to degraded inputs. VLM evaluation and human preferences on real photographs further support identity preservation and degradation correction. More demos and code are available at https://supra-3d.github.io/gate3dgen.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Object (grammar), Computer science, Artificial intelligence, Engineering, Computer vision, Fidelity

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange